��<!DOCTYPE html> <html lang="en"> <head> <meta charset="UTF-8"> <meta name="viewport" content="width=device-width, initial-scale=1.0"> <title>Dynamic Spatial Audio Soundscapes with Latent Diffusion Models</title> <style> body { font-family: Arial, sans-serif; margin: 2em; line-height: 1.6; } header, footer { text-align: center; margin-bottom: 2em; } nav a { margin: 0 1em; color: #3498db; text-decoration: none; } section { margin-bottom: 3em; } .demo-grid { display: grid; grid-template-columns: repeat(2, 1fr); gap: 10em; justify-items: center; } .demo { text-align: center; } .prompt { font-weight: bold; margin-bottom: 0.5em; } .demo iframe { display: block; margin: 0 auto; max-width: 100%; height: 100%; } .abstract { max-width: 700px; margin: 0 auto 1em auto; text-align: center; } </style> </head> <body> <header> <h1>Dynamic Spatial Audio Soundscapes with Latent Diffusion Models</h1> <p><em>Christian Templin, Hao Wang</em></p> <p><em>Stevens Institute of Technology</em></p> <p class="abstract"><strong>Abstract:</strong> Spatial audio is an integral part of immersive entertainment, such as VR/AR, and has seen increasing popularity in cinema and music as well. The most common format of spatial audio is described as first-order Ambisonics (FOA). We seek to extend recent advancements in FOA generative AI models to enable the generation of 3D scenes with dynamic sound sources. Our proposed end-to-end model comes in two variations which vary in their user input and level of precision in sound source localization. In addition to our model, we also present a new dataset of simulated spatial audio-caption pairs. Evaluation of our models demonstrate that they are capable of matching the semantic alignment and audio quality of state of the art models while capturing the desired spatial attributes.</p> <nav> <a href="LINK_TO_PAPER" target="_blank">Paper</a> <a href="LINK_TO_CODE_REPO" target="_blank">Code Repository</a> <a href="LINK_TO_CHECKPOINT_A" target="_blank">Descriptive Model Checkpoint</a> <a href="LINK_TO_CHECKPOINT_B" target="_blank">Parametric Model Checkpoint</a> </nav> </header> <main> <section id="model-a"> <h2 class="demo">Descriptive Model Demos</h2> <div class="demo-grid"> <div class="demo"> <p class="prompt">Prompt: "A police siren wails from the front and slowly moves counter-clockwise to the back"</p> <iframe width="560" height="315" src="https://www.youtube.com/embed/yOE3MLrbyhY?si=rnLB0dnXJ8tBb34l" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe> <a class="download-link" href="samples\descSIREN.wav" download>Download WAV</a> </div> <div class="demo"> <p class="prompt">Prompt: "A lion roars from the back right and below"</p> <iframe width="560" height="315" src="https://www.youtube.com/embed/IQ3g9HW7Lzs?si=R4NBnNwqJijKD4Wj" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe> <a class="download-link" href="samples\descROAR.wav" download>Download WAV</a> </div> <div class="demo"> <p class="prompt">Prompt: "A laughing man slowly moves counter-clockwise from the left to the front"</p> <iframe width="560" height="315" src="https://www.youtube.com/embed/j6V_I6uZiVw?si=u18qMIJWjG__dd1B" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe> <a class="download-link" href="samples\descLAUGH.wav" download>Download WAV</a> </div> <div class="demo"> <p class="prompt">Prompt: "A bird chrips from the front and above, then flies below"</p> <iframe width="560" height="315" src="https://www.youtube.com/embed/1_22KX-V97s?si=H4QnSpTC-sTPln3h" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe> <a class="download-link" href="samples\descBIRD.wav" download>Download WAV</a> </div> </div> <!-- Add more demos for Model A as needed --> </section> <!-- <section id="model-b"> <h2 class="demo">Parametric Model Demos</h2> <div class="demo-grid"> <div class="demo"> <p class="prompt">Prompt: "Your prompt here for Model B"</p> <iframe width="560" height="315" src="https://www.youtube.com/embed/VIDEO_ID_B1" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> </div> <div class="demo"> <p class="prompt">Prompt: "Your prompt here for Model B"</p> <iframe width="560" height="315" src="https://www.youtube.com/embed/VIDEO_ID_B1" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> </div> <div class="demo"> <p class="prompt">Prompt: "Your prompt here for Model B"</p> <iframe width="560" height="315" src="https://www.youtube.com/embed/VIDEO_ID_B1" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> </div> <div class="demo"> <p class="prompt">Prompt: "Your prompt here for Model B"</p> <iframe width="560" height="315" src="https://www.youtube.com/embed/VIDEO_ID_B1" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> </div> </div> </section> --> </main> <footer> <!-- <p>&copy; 2025 Your Name or Institution</p> --> </footer> </body> </html>